Process and system for identifying damage

Abstract

A process and system to provide damage identification and assessment of damage to a geographic area may include acquiring imagery data of a geographic area, processing the imagery data using wavelet transformation to identify damage to the geographic area and outputting a map showing damage condition of the geographic area. Processing the imagery data may use wavelet transformation that outputs wavelet transformation images. Damage categories for at least one location in the imagery data may be provided using discriminant analysis applied to the wavelet transformation images. The outputted maps and damage categories may be used to assess damage to areas affected by catastrophic-like events such as, e.g., hurricanes, floods, earthquakes, tornadoes and the like. This process is faster and may be more accurate than current assessment techniques thereby permitting quick responses to catastrophic-like events.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS[0001]This application claims benefit and priority to U.S. Provisional Application No. 61 / 734,685 filed on Dec. 7, 2012, the disclosure of which is incorporated by reference herein in its entirety, including Appendices and Attachments.BACKGROUND OF THE INVENTION[0002]1.0 Field of the Disclosure[0003]The present disclosure relates to a process, computer program product and system of image processing to identify, map and / or classify damage of geographic areas and, more particularly, the present disclosure relates to a process, computer program product and system of image processing using wavelet transformation to identify, map and / or classify damage of geographic areas, such as damage caused by, e.g., earthquakes, hurricanes, tornadoes, floods and the like.[0004]2.0 Related Art[0005]Hurricanes on average make landfall on U.S. soil 1.2 (for El Niño years) to 2.1 times per year (for La Niña years) resulting in normalized mean damage of $7.7 (El Niñ...

AI Technical Summary

Benefits of technology

[0011]According to one non-limiting example of the disclosure, a system, a process and computer program product is provided that identifies and assesses damage to a targeted geographic area using imagery data that provides results very fast compared with prior traditional techniques. The identification and assessment may include processing image data of a target geographic area using wavelet transformation. Also, applying a discriminant analysis to the wavelet transformation output may provide a damage category for at least one location in the image data of the target geographic location.

Problems solved by technology

There are several potential issues with a visual interpretation based damage assessment approach that may be addressed by use of an algorithmic image damage assessment approach.

Rapid response visual interpretation of overhead imagery requires significant human and infrastructure resources.

Hurricanes typically cause wide areas of damage resulting in large volumes of imagery.

Second, damaging hurricanes make landfall at unpredictable times and with uncertain periodicity.

This sporadic and unpredictable nature of hurricane landfall hinders the development and maintenance of a dedicated, trained, ongoing human capability for visual damage assessment.

This ad hoc human capability makes it difficult for damage assessments to be performed consistently across large storms and from storm to storm.

Fourth, a human visual interpreter based method for damage assessment is not easily calibrated.

Moreover, past systems and techniques have been slow (often providing results in terms of days or weeks) and at times inaccurate to identify and assess damage to target areas.

Often human interpretation is required of current type systems that may lead to inaccurate assessments.

Images